A Comparative Analysis of Data-Driven Consolidation Policies for Energy-Efficient Clouds

Consolidation of virtual machines is one of the key strategies used to reduce the power consumption of Cloud data centers. For this reason it is extensively studied. Consolidation has the goal of allocating virtual machines on a few physical servers as possible, while satisfying the Service Level Agreement established with users. Nevertheless, the effectiveness of a consolidation strategy strongly depends on the forecast of virtual machine resource needs. To this aim, data-driven predictive models can be exploited to develop intelligent consolidation policies. This paper describes a comparative analysis of consolidation strategies of virtual machines in Cloud systems, driven by predictive data mining models. In particular, migrations are driven by the forecast of the future computational needs (CPU, RAM) of each virtual machine, in order to efficiently allocate those on the available servers. Experimental results, performed on data of a real Cloud data center, shows several insights in terms of energy saving and most efficient consolidation strategies.